I couldn't find much help on the internet, which is why I'm asking this here.
Spyder has an issue where input() will cause an issue to the app. I've heard that spyder 5.3.3 has this issue fixed.
I have done:
conda update conda
conda update anaconda
However, when I try conda install spyder=5.3.3, it says failed with initial frozen solve. Retrying with flexible solve.
It then never solves the environment. Any help?
I tried all the solution posted but they does not work for my problem. I finally solved it by first uninstalling the existing spyder version and installing a new one. To do this, use this:
conda uninstall spyder
conda install spyder = 5.3.3
I don't have a great solution, ran into the same issue myself, I was able to install version 5.2.2 where 5.1.5 is the base version. This version does also fix the input() problem you were having.
Alternatively you might try either of these two options to install spyder into it's own environment which should negate the problems but will require you to be in that environment to run spyder.
conda create -n spyder-env spyder
or
conda create -n spyder-env spyder numpy scipy pandas matplotlib sympy cython
Close spyder. Run command: pip install --upgrade spyder
worked for me.
Me as well. At least you could get it updated to 5.2.2 using the recommended update command from anaconda's page for spyder:
conda install -c anaconda spyder
Or use Vaxion's solution by creating an environment.
I had the same problem, I tried on many ways. But this last worked for me:
conda install -c anaconda spyder
Maybe you can try it too.
I took it from anaconda / packages / spyder 5.3.3
Probably your Environment is broken somehow. I suggest you to create a new environment specifying conda-forge as a channel already at creation time:
conda create -n spyder-env -c conda-forge python=3.10 spyder=5.3.3
The newest versions of Spyder are usually available on this channel. Then you can install your other packages and libraries there as well.
Like others, the upgrade was advancing very slowly. It was not urgent, so I just left it, and it took about a week (!) and ended by barfing out 7000+ lines of package conflicts.
But then I followed the advice from Qiyuan Chen and it worked in just a few minutes:
conda uninstall spyder
conda install spyder=5.3.3
Qiyuan Chen's answer also worked for me. After I uninstalled and reinstalled it, I went back to Anaconda Navigator and the Spyder icon was there with an install button underneath it. I hit 'INSTALL' and it worked.
As per others and Peters answer above
But then I followed the advice from Qiyuan Chen and it worked in just a few minutes:
Needed to turn the Anaconda Navigator off and on againbut I've now got Spyder 5.3.3 up and going with no distracting messages
Related
I'm trying to update Spyder to version 4.0.0b2 but I'm getting an error. I don't have a lot of experience with cmd or anaconda prompt, so I don't understand the problem. Any ideas?
I'm not on MS Windows but I see several things in your screenshot
You are on base environment
You already have spyder 3.3.3 and that means that you have to use conda update spyder or uninstall it and then install it with your command conda install -c spyder-ide spyder=4.0.0b2
The same with all remaining packages mentioned there (preferably one by one since some may disappear like the kernel one when you uninstall spyder 3.3.3)
If you see that Anaconda cannot find the package in your current channels, try via proposed channels on Anaconda Cloud like conda search --override-channels --channel URL/PACKAGE
Hope it helps (:
I'm trying to create a Python 2.7 virtual environment with Anaconda so I can download some packages that are not compatible with Python 3.4. I'm working in Conda version 4.2.13.
When I type the command conda create -n chemistry python=2.7 anaconda the whole thing freezes at the solving package specifications stage.
Does anyone know what causes this or how I can go about fixing it?
Today I faced the same snag. It got fixed after updating my Anaconda Navigator.
Updating your Anaconda Navigator may fix your issue, too.
Try this:
conda create -n chemistry python=2.7
The initial command you use will try to install a package named anaconda. If you want to specify the channel when installing packages,you can add -c <channel>.For example:
conda install -c conda-forge tqdm
Had same issue while I was trying to install some packages. I tried updating python then all seemed working. Try it out
conda install python
Python 3.7 alpha version is out, but I haven't been able to find any post on how to update to python 3.7 using Anaconda - maybe they will wait for the official release? Any suggestions?
This can be installed via conda with the command conda install -c anaconda python=3.7 as per https://anaconda.org/anaconda/python.
Though not all packages support 3.7 yet, running conda update --all may resolve some dependency failures.
Python 3.7 is now available to be installed, but many packages have not been updated yet. As noted by another answer here, there is a GitHub issue tracking the progress of Anaconda building all the updated packages.
Until someone creates a conda package for Python 3.7, you can't install it. Unfortunately, something like 3500 packages show up in a search for "python" on Anaconda.org (https://anaconda.org/search?q=%22python%22) so I couldn't see if anyone has done that yet.
You might be able to build your own package, depending on what OS you want it for. You can start with the recipe that conda-forge uses to build Python: https://github.com/conda-forge/python-feedstock/
In the past, I think Continuum have generally waited until a stable release to push out packages for new Pythons, but I don't work there, so I don't know what their actual policy is.
To see just the Python releases, do conda search --full-name python.
The September 4th release for 3.7 recommends the following:
conda install python=3.7 anaconda=custom
If you want to create a new environment, they recommend:
conda create -n example_env numpy scipy pandas scikit-learn notebook
anaconda-navigator
conda activate example_env
run conda navigator, you can upgrade your packages easily in the friendly GUI
conda create -n py37 -c anaconda anaconda=5.3
seems to be working.
I have recently Installed Anaconda for Python 3.6 but it shows the error "Segmentation fault" whenever I try to run Anaconda-Navigator.
I've tried just writting in the terminal Anaconda-Navigator and also going to my Anaconda3 folder and try to execute it inside bin.
The only solution that works so far is accessing the previously bin folder as root. My problem is that I need to activate TensorFlow before I run anything in my console, but that is imposible as a root user.
I've already try to upgrade both, Anaconda and Navigator and reinstall them but nothing ocurrs
Anyone here has any idea of what is happening?
Had the same problem with my anaconda install. It is probably related to incompatible versions of conda and anaconda navigator.
conda update --all
worked for me.
I had the same issue today, and I solved by running:
conda upgrade anaconda
The output said:
The following packages will be DOWNGRADED due to dependency conflicts:
anaconda-navigator: 1.5.1-py36_0 --> 1.5.0-py36_0
after that, I launched anaconda-navigator (no segfault this time) and a window prompted me to upgrade to 1.5.1. I clicked yes, and now running anaconda-navigator 1.5.1 without segfaults on launch. Don't know what is the reason
I tried to install opencv via conda then I got the same error. Later I followed the
conda update conda
conda update anaconda-navigator
conda update navigator-updater
Then I did
conda update --all
Voila, its working again
I had the same issue when I install OpenCV library using conda.Most probably downgrading something makes this issue happen. Just type :
conda update --all
I had the same problem.I solved it by adding /lib to mt LD_LIBRARY_PATH.
Note: On my system Anaconda installation path is /home/pushyamik/anaconda3.
What is a proven method for installing pydotplus for Python 3.5 on a 64-bit Windows(10) system? So far I haven't had any luck using conda or a number of other approaches.
It appears there are several viable options for both Linux Ubuntu and Windows for Python 2.7. Unfortunately it's necessary for me to use this particular configuration, so any suggestions would be greatly appreciated!
Try running anaconda prompt as 'administrator', then use:
conda install -c conda-forge pydotplus
What have you tried for conda? According to https://anaconda.org/conda-forge/pydotplus/files pydotplus is available from conda-forge for Windows x64 and Python 3.5. Typing
conda install -c conda-forge pydotplus
should work.
I had definitely tried that earlier but for some reason it wasn't working. However, thanks to your response I took another look at my configuration to see why that particular install package wasn't running properly.
I originally used Anaconda to install a dual Python 2.7/3.5 environment and while I was always able to run 3.5 without any issues in an IDE (Jupyter Notebook in this instance), the current environment had still automatically defaulted to 2.7. The quick way to check for this via the command window is:
C:\Users\Username\Anaconda\envs>conda env list
which should produce something like:
# conda environments:
#
py27 * C:\Users\Username\Anaconda\envs\py27
py35 C:\Users\Username\Anaconda\envs\py35
Note that there will be '*' next to the environment that is currently active. To switch, simply type 'activate' followed by the new desired environment name and the change should immediately be verified by the command prompt that is returned:
C:\Users\Username\Anaconda\envs>activate py35
(py35) C:\Users\Username\Anaconda\envs>
After making the environment change, I reran the conda install package for pydotplus that you mentioned:
conda install -c conda-forge pydotplus
and it worked perfectly!
Two other quick things to consider when installing pydotplus to ensure that it interacts correctly with GraphViz in your particular environment:
Make sure you're running the most up to date version of scikit-learn. More info is provided on the official website (http://scikit-learn.org/stable/install.html), including the quick update command: conda update scikit-learn
Verify that the GraphViz bin directory has been added to your User Path: (Why is pydot unable to find GraphViz's executables in Windows 8?)
I had the same issue. here what I did was,
first I ran the anaconda prompt as administrator and then give the command,
**conda install -c conda-forge pydotplus**
it worked for me well